Takeovers and Mergers - Case Study Example

Analysis of the non frustration principle of the city code on takeovers and mergers Structural takeover defenses--poison pills and shark repellents [1] such as staggered boards--have long been controversial. At the core οf the controversy is a striking split in how defenses are viewed by legal academics, on the one hand, and practicing lawyers, judges, and legislators, on the other hand. Exceptions exist, but academics generally have taken a dim view οf takeover defenses, and practicing lawyers have generally supported defenses in advising clients, with judges and legislators siding more with practitioners than with academics. Academic hostility to defenses has been built in large part on the common belief that defenses reduce firm value, a view thought supported by both economic theory and empirical evidence. The theory οf how defenses reduce firm value is a simple application οf agency cost analysis: agency costs make defense adoption possible and likely, and defenses in turn increase agency costs by making it harder for principals (shareholders) to replace or otherwise discipline agents (directors) through a takeover. [2] But the academic conviction that agency cost theory is the lens through which to view takeover defenses has been reinforced by empirical studies. Legal academics widely believe that those studies show that firms' stock prices fall on average when firms adopt defenses. Without that evidence, the theoretical case against defenses remains, but is much less compelling, particularly for policy making. Practitioner support for defenses no doubt stems in part from the fact that defense adoption (and litigation over defenses) provides legal practitioners with profits. But practitioners have also looked to economic theory and empirical evidence for support in convincing boards οf directors that they are justified in adopting defenses and in persuading lawmakers not to intervene against defenses. The evidence in favor οf defenses has been produced for the most part not by academics but by investment banks and proxy solicitors, and it shows that defenses, such as pills, increase the premiums target shareholders receive in takeovers. This evidence is consistent with the theory that well-motivated (or adequately constrained) boards use defenses not to entrench themselves or defeat advantageous bids, but to seek better alternatives or bargain for target shareholders and extract a greater share οf deal synergies than they otherwise could do. Recently, both academics and practitioners have been confronted with a new source οf evidence on takeover defenses, and the results are decidedly mixed, supporting neither group's view with certainty. Several recent and ongoing studies show that prior to initial public offerings (IPOs), a significant number οf firms adopt terms making takeovers more difficult than does default law, [3] which seems to fly in the face οf the academic belief that defenses reduce firm value. [4] Even sophisticated pre-IPO shareholders (such as venture capitalists and leveraged buyout firms) permit their adoption, [5] and firms with defenses are more likely to be represented by higher-quality investment banks in an IPO. At the same time, institutional investors routinely oppose proposals by firms to adopt defenses midstream, and studies οf IPOs show that defenses vary significantly, contrary to the beliefs οf legal practitioners that a full set οf defenses is privately optimal for all firms. These surprising and mixed results make a reassessment οf prior empirical evidence on takeover defenses worthwhile, both to examine the methods used and to assess the strength οf support the evidence provides for the opposing academic and practitioner positions on takeover defenses. A better understanding οf defenses is given some urgency by the strength οf the market for corporate control in the 1990s. Despite widespread adoption οf defenses, nearly seventy hostile takeover bids were made in 1995, well over the average annual number οf bids in the 1980s, and almost eighty percent οf the number made in 1988, the peak year for hostile activity in the 1980s. [6] Negotiated deals--including many that would not have been done but for the background threat οf a hostile bid or boardroom coup--have broken records in each οf the past five years, reaching an all-time high in the United States in 1999, even after accounting for growth in the overall economy. At a minimum, the roaring mergers and acquisitions (M&A) market shows that defenses are not major impediments to the movement οf assets to "higher valuing users" via the market for corporate control. The main task οf this Article is to survey twenty years οf "scientific evidence" on poison pills and other defenses. Somewhat surprisingly, given that financial economists have produced over fifty empirical studies in that time, no such survey exists in either the legal or the economic literature. The more specific goals οf this Article are to provide a more careful analysis οf the way defenses work in practice, with a special focus on the way they interact, and to present new evidence to support that analysis where it departs from the analysis implicit in prior empirical work. As will be shown, and contrary to conventional views οf either academics or practitioners, the scientific evidence to date does not provide strong support either for opposing or supporting defenses because it has been uninformed by a sufficiently fine-grained theory οf how defenses work in practice. Together, the critical survey, practical analysis, and new evidence have important implications for future research, practical advice, and policy. For researchers, tasks to be done are many, varied, and detailed, but the research agenda should be modified in two general ways: first, focus less on midstream defense adoptions and more on pre-IPO defenses; and second, recognize that defense interactions are often more important than the effect οf any one defense in isolation. For practitioners, the good news is that the conventional advice for firms to adopt defenses remains unindicted by the scientific evidence, but the bad news is that research also provides no evidence for boards to rely upon in adopting defenses, for litigators to use in litigation, or for corporate lawyers to influence institutional investors (which remain generally opposed to defenses) or lawmakers. For courts and legislators, the lessons are cautionary: while lawmakers were correct not to swallow academic anti-defense prescriptions hook, line, and sinker, they should remain open-minded about defenses and continue to pay attention to theoretical and empirical analysis οf defenses as it develops. Until recently, empirical research on takeover defenses has been dominated by "event studies" οf poison pills and midstream shark repellents. It is fair to say that event studies have provided the principal evidence supporting legal academic views οf the effects οf defenses on shareholder wealth and social welfare. [7] Easterbrook and Fischel rely on such studies as the primary evidence for asserting that "[e]very device giving managers the power to delay or prevent an acquisition makes shareholders worse off." [8] Romano states that "event studies οf defensive tactics find significant negative returns on their adoption" [9] and cites those studies to support the statement that poison pills are "most likely to defeat [takeover] bids and therefore to diminish shareholder wealth . . . ." [10] Event study methodology is well-known and generally accepted as providing useful information about the wealth effects οf legal and other events affecting stock prices. The relationship between a firm's stock price and the overall stock market is estimated, that estimate is used to predict price changes during a specified interval οf time that includes the event being studied, and the differences between predicted and actual returns during the event interval are summed, producing a "cumulative abnormal return" (CAR). Event studies are premised on the assumption that stock prices are unbiased estimates οf firm value--that is, even if prices are inaccurate, they are off by an amount that averages zero in large samples. Researchers thus calculate CARs from many similar events, test the average effects for statistical significance, and state how likely it is that market reactions were caused by information about the event, rather than unspecified causes or random chance. Event studies are subject to a number οf potential methodological flaws, many οf which have not been addressed in studies οf defenses. But even at face value, event studies οf takeover defenses produced little reliable evidence either way on the wealth effects οf takeover defenses. Within a given study, results are mixed and weak; between studies, results are inconsistent; over time, results have become less significant (both statistically and economically); and when firms are partitioned on various traits, results differ among subsamples. Even with no further analysis, event studies do not provide much if any support for theoretical (positive) arguments that defenses harm shareholders, or for normative arguments that defenses should be prohibited. Nor do they provide any assistance in understanding how defenses might improve firm value, or improve firm value at some firms and not at others, and so cast little light on why some but not all firms adopt defenses prior to IPOs. The most well-known and frequently cited empirical studies οf takeover defenses take as their object οf study the adoption οf a poison pill. Jarrell and Ryngaert led the way with a 1986 study οf 245 pills, published with the imprimatur οf the Securities and Exchange Commission's (SEC) Office οf the Chief Economist. [11] Yet their general results were that in the two days following adoption, stock prices οf adopting companies fell on average, net οf market movements, only by an absolutely small amount not statistically different from zero (0.22%). Even after the authors focused on a relatively tiny subsample (n = 15), choosing pill adoptions in ways that appear ad hoc and may well have biased their results, they were only able to find a relatively modest impact (-2.21%). [12] By comparison, merger premiums averaged 42% over pre-bid market prices in the 1980s, and premiums in hostile takeovers were typically larger than premiums in negotiated deals. Not every firm adopting a pill would receive such a premium price, but the odds should have been high for firms in their sample, which the authors had limited to only those firms subject to takeover speculation. If pills substantially impaired the odds that target shareholders would receive 50% premiums, a price decline at a carefully selected group οf likely targets should have been significantly greater than 2.21%. In sum, results οf early event studies οf poison pills were statistically mixed and economically weak. The studies suggest that even if the samples used were representative, the wealth effects οf pills were neither large, nor certain, nor general. Nevertheless, the authors felt able to conclude that "poison pills are harmful to target shareholders, on net," [13] a mischaracterization (or at least an exaggeration) common to the early pill studies that has been parroted ever since. Despite economically weak and statistically nonrobust results, the Jarrell and Ryngaert study attracted widespread attention and acceptance, and continues to be cited as proοf that defenses reduce shareholder wealth. [14] Since 1986, the majority οf poison pill event studies have followed Jarrell and Ryngaert in attempting to resolve the debate over pills' wealth effects. Subsequent studies have produced results that make the case against pills look even weaker. The majority οf studies summarized in Appendix A show no significant impact on price for the full samples. If results from full samples in all studies using two- or three-day event intervals are pooled, the weighted average price reaction is +0.02%. In other words, the net price impact οf pill adoptions has been positive, albeit close to zero. The net results are two orders οf magnitude smaller than two- or three-day price effects οf secondary stock offerings (-3.0%), [15] announcements οf acquisitions (abnormal returns ranging from-1.2% to -3.3%), spin-offs (+3.4%),[160] deaths οf inside 5% or greater blockholders (+3.01%), [17] and sales οf 5% or greater blocks οf stock (+5.1%).[18] The net results are much smaller even than effects οf nonbinding agreements to make relatively minor governance reforms, such as the adoption οf confidential voting (+0.9234%). [19] Needless to say, legal academics have not taken nearly as strong a position, if any, on these other events as they have on defenses, despite stronger evidence suggesting such events have more effect on shareholder wealth. Legal academics have not, for example, suggested a mandatory dispersion rule for large blockholders, despite evidence suggesting such a rule would on average have more than double the effect οf a rule against pills. Not only are the overall results quite weak, but they have been inconsistent over time. Studies οf early pill adoptions show (weak) negative results, whereas the only studies οf pill adoptions after 1986 show no statistically significant results for their full samples. Comment and Schwert's recent analysis, which uses the largest sample οf any οf the studies (n = 1459), produces no statistically significant results for adoptions in any year except 1984, prior to judicial approval and widespread adoption οf the pill. The contrasts between results over time might suggest that investors simply misestimated early pills' effects--the pill, after all, was a true innovation. But the inconsistencies are too serious and general to be attributed to different temporal samples. Studies that report results over varying event intervals reveal that reactions are sensitive to the event interval chosen. Results are also inconsistent between studies that examine adoptions in the same period. For example, studies by Malatesta and Walkling and by Choi et al. produced small but statistically significant negative results for their full sample, [20] which contrasts with findings οf no significant effects by Jarrell and Ryngaert and by Ryngaert in his larger follow-up study, despite the fact that each οf these studies examined pill adoptions in the same time period (1982-1986). Each early study, after excluding firms with "confounding events," found weak negative effects in subsamples οf pills adopted in the mid-1980s, but Brickley, Coles, and Terry, also after excluding confounding events, find no significant effects for pill adoptions in the same period. Together, these inconsistencies suggest a more serious problem with using event study methodology to capture pills' wealth effects. More recent commentators have used event studies οf pills to contrast wealth effects on different partitions οf firms adopting pills based on (and So test propositions about the importance and role of) independent directors, insider trading, or institutional shareholders. This shift in research focus has not followed any significant reconsideration οf the event study methodology as applied to defenses and may instead be attributable to many scholars having read the conclusion and not the details οf the earlier studies (and so thought the basic research question οf pill wealth effects was settled). Alternatively, scholars may have shifted research focus because the legality οf the pill has been settled in nearly all states [21] --making the policy debate less urgent or interesting, or because the number οf bids fell dramatically--if briefly--in the early 1990s, or simply out οf a desire to say something new. Given the dominant academic theory that pills represent (and are enabled by) a form οf agency cost, the weak and inconsistent findings οf poison pill event studies are puzzling. Why don't pill adoptions reduce share prices? One answer may be that agency cost theory is not the whole story. But a more straightforward answer is that pill event studies suffer from a serious design flaw, which renders them more problematic for positive research than traditional critiques might suggest, and are οf no use for policy analysis. In brief, pill adoption rarely has any real effect on the legal takeover vulnerability οf the firm adopting a pill. To see why, imagine you are a bidder, considering a high premium bid for a target. Imagine the target lacks a pill. Do you conclude that the target is easy to acquire, as a matter οf law? Clearly not. Unless there is something unusual about the target, the target will have the ability to--and almost certainly will--adopt a pill as soon as you start your bid. For large, sophisticated targets, pill adoption can occur in a single business day: the only legal action necessary is a board meeting and approval, lawyers can keep necessary documents at the ready, and directors can meet by conference call on a few hours notice. Even for less sophisticated firms, takeover bids are subject to sufficient delay under the Williams Act and the Hart-Scott-Rodino Antitrust Improvements Act οf 1976 ("HSR Act") that a target will rarely if ever be prejudiced by failing to adopt a pill in advance. Numerous targets have in fact adopted pills in response to bids, and evidenceshows that targets do not allow themselves to be taken over solely because they had not adopted pre-bid pills. Notes (1.) Charter and bylaw provisions that impede hostile bids are colloquially known as "shark repellents." See generally Ronald J. Gilson, The Case Against Shark Repellent Amendments: Structural Limitations on the Enabling Concept, 34 STAN. L. REV. 775 (1982). (2.) See, e.g., RONALD CHARLES CLARK, CORPORATE LAW 589 (1986) (asserting that "adoption οf takeover defensive tactics usually involves extremely severe self-dealing"); FRANK H. EASTERBROOK & DANIEL R. FISCHEL, THE ECONOMIC STRUCTURE ΟF CORPORATE LAW 168-74 (1991) (setting out an "agency cost model," favoring hostile takeovers, and criticizing defenses). (3.) John C. Coates IV, Explaining Variation in Takeover Defenses: Failure in the Corporate Law Market (June 26, 2000) (4.) EASTERBROOK & FISCHEL, supra note 5, at 167 ("[I]f [defenses] are not beneficial to investors, they will depress the price οf the stocks affected . . . [which] will be more attractive as takeover targets. . . . In the long run, useful provisions will dominate."). (5.) Daines & Klausner, supra note 8, at 1, 13-14. (6.) See Coates, How Contestable?, supra note 13, at 855. (7.) See supra note 6 and infra notes 36-39. Many economists have also been convinced by event studies. (8.) EASTERBROOK & FISCHEL, supra note 5, at 204 (emphasis in original); see also id. at 196-98 (summarizing event studies listed in id. at 209-11); id. at 205 (citing "the absence οf any existing [takeover defense] that increases targets' market value" after reviewing event studies) (emphasis in original). (9.) ROMANO, GENIUS, supra note 6, at 70-71 (comparing defense studies with state takeover statute studies). (10.) Id. at 80 & n.58 (contrasting the purportedly less negative effects οf golden parachutes and greenmail with the effects οf pills) (emphasis added). (11.) Gregg Jarrell & Michael Ryngaert, Office οf Chief Economist οf the Securities and Exchange Commission, The Effects οf Poison Pills on the Wealth οf Target Shareholders (Oct. 23, 1986) (on file with author). (12.) Jarrell & Ryngaert, supra note 27. Other studies found no results in their full samples and weak results in much smaller subsamples. See, e.g., Robert Comment & G. William Schwert, Poison or Placebo? Evidence on the Deterrence and Wealth Effects οf Modern Antitakeover Measures, 39 J. FIN. ECON. 3, 18-20 (1995); Ryngaert, supra note 29, at 391-92. (13.) Jarrell & Ryngaert, supra note 27, at 43. Their conclusion was also based on a study οf 30 takeover battles involving pills in which 45% οf firms remained independent, resulting in short-term price declines, and in which another 45% οf firms were acquired at higher prices in auctions. Id. at 4. Net, targets experienced a net-of-market return over six months οf -2.0%. Id. at 25-28. These results do not support a strong or general position against pills. (14.) See, e.g., EASTERBROOK & FISCHEL, supra note 5, at 210; Matthew Garms, Shareholder ByLaw Amendments and the Poison Pill: The Market for Corporate Control and Economic:' Efficiency, 24 J. CORP. L. 433, 449 n.172 (1999) (citing Jarrell & Ryngaert, supra note 27, and Malatesta & Walkling, supra note 32, as "supportive οf a conclusion that the adoption οf poison pills has a negative effect on shareholder wealth"). (15.) See RICHARD A. BREALEY & STEWART C. MYERS, PRINCIPLES ΟF CORPORATE FINANCE 349 (4th ed. 1991) (summarizing results from three studies). (16.) Lane Daley, Vikas Mehrotra & Ranjini Sivakumar, Corporate Focus and Value Creation: Evidence From Spinoffs, 45 J. FIN. ECON. 257, 265 tbl.2 (1997). (17.) Myron B. Slovin & Marie E. Sushka, Ownership Concentration, Corporate Control Activity, and Firm Value: Evidence from the Death οf Inside Blockholders, 48 J. FIN. 1293, 1300 tbl. II (1993). (18.) Michael J. Barclay & Clifford G. Holderness, Negotiated Block Trades and Corporate Control, 46 J. FIN. 861, 866 tbl. II (1991). (19.) Deon Strickland, Kenneth W. Wiles & Marc Zenner, A Requiem for the USA: Is Small Shareholder Monitoring Effective?, 40 J. FIN. ECON. 319, 334 tbl.5 (1996). (20.) Choi, Kamma & Weintrop, supra note 53, at 383 tbl. 1; Malatesta & Walkling, Poison Pill Securities, supra note 32, at 360-66. (21.) See Robinson, Coates & Presser, supra note 52 (showing that in all states in which state courts found poison pills illegal, state legislatures had overturned the decisions). Read More